Apertus 70B vs GLM-4.5V vs Qwen3-Next 80B-A3B Instruct
GLM-4.5V comes out ahead, 49 to 39 and 33 on our weighted score.
Swiss AI
Apertus 70B
33/100- ECI—
- Price$0.82 / $2.42
- Context66K
- Our pick
Z.ai (Zhipu)GLM-4.5V
49/100- ECI—
- Price$0.60 / $1.80
- Context64K
Alibaba (Qwen)Qwen3-Next 80B-A3B Instruct
39/100- ECI—
- Price$0.50 / $2.00
- Context131K
GLM-4.5V is our pick
GLM-4.5V is the better all-round choice, scoring 49/100 against Qwen3-Next 80B-A3B Instruct (39) and Apertus 70B (33). It leads on inputs & features. Qwen3-Next 80B-A3B Instruct wins on context window. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
- CapabilityNot enough dataNo independent benchmark covers every model here yet
- Lowest priceQwen3-Next 80B-A3B InstructQwen3-Next 80B-A3B Instruct $0.875 · GLM-4.5V $0.90 · Apertus 70B $1.22 per 1M tokens (3:1 blend)
- Longest contextQwen3-Next 80B-A3B InstructQwen3-Next 80B-A3B Instruct 131,072 · Apertus 70B 65,536 · GLM-4.5V 64,000 tokens
- Widest inputsGLM-4.5VApertus 70B: Text · GLM-4.5V: Text, Images, Video · Qwen3-Next 80B-A3B Instruct: Text
- Self-hostingAll open weightsEvery model here can be downloaded and run on your own hardware
| Measure | Weight | Apertus 70B | GLM-4.5V | Qwen3-Next 80B-A3B Instruct |
|---|---|---|---|---|
| Price | 50% | 46 | 52 | 53 |
| Inputs & features | 30% | 25 | 70 | 25 |
| Context window | 20% | 12 | 12 | 24 |
| Overall | 100% | 33/100 | 49/100 | 39/100 |
Left out because at least one model lacks the data: capability. The remaining weights were rescaled.
Every spec in one table
Highlighted cells lead their row. Dashes mean the data is not published.
| Specification | Apertus 70BSwiss AI | GLM-4.5VZ.ai (Zhipu) | Qwen3-Next 80B-A3B InstructAlibaba (Qwen) |
|---|---|---|---|
| Capability | |||
| Capabilities Index (ECI) | — | — | — |
| ECI rank | — | — | — |
| Price per million tokens | |||
| Input | $0.82 | $0.60 | $0.50 (best) |
| Output | $2.42 | $1.80 (best) | $2.00 |
| Cached input | — | — | — |
| Blended (3:1) | $1.22 | $0.90 | $0.875 (best) |
| Long-context rate | Same rate | Same rate | Same rate |
| Price source | Median of 3 providers | Official Z.AI API | Official Alibaba API |
| Limits | |||
| Context window | 65,536 tokens | 64,000 tokens | 131,072 tokens (best) |
| Max output | 8,192 tokens | 16,384 tokens | 32,768 tokens (best) |
| Inputs and features | |||
| Text | Yes | Yes | Yes |
| Images | No | Yes | No |
| PDFs | No | No | No |
| Audio | No | No | No |
| Video | No | Yes | No |
| Reasoning | No | Yes | No |
| Tool calling | Yes | Yes | Yes |
| Structured output | No | No | No |
| Availability | |||
| Weights | OpenApache-2.0 | Open | Open |
| API model ID | — | glm-4.5v | qwen3-next-80b-a3b-instruct |
| API providers | 3 | 11 | 13 (best) |
| Released | Sep 2, 2025 | Aug 11, 2025 | Sep 2025 |
| Knowledge cutoff | Sep 2025 | Apr 2025 | Apr 2025 |
What would a month cost?
Enter your expected volume in millions of tokens. List prices only; caching and batch discounts would lower these.
- Apertus 70B$13.04
- GLM-4.5V$9.60
- Qwen3-Next 80B-A3B Instruct$9.00
Which should you choose?
Which is better: Apertus 70B, GLM-4.5V or Qwen3-Next 80B-A3B Instruct?
GLM-4.5V is the better all-round choice, scoring 49/100 against Qwen3-Next 80B-A3B Instruct (39) and Apertus 70B (33). It leads on inputs & features. Qwen3-Next 80B-A3B Instruct wins on context window. The score weighs price 50%, inputs & features 30%, context window 20%. None of the shared benchmarks cover every model here yet, so capability is left out of this verdict.
Which is cheaper, Apertus 70B, GLM-4.5V or Qwen3-Next 80B-A3B Instruct?
Qwen3-Next 80B-A3B Instruct is cheaper at $0.50 input / $2.00 output per million tokens (official Alibaba API price). GLM-4.5V costs $0.60 input / $1.80 output per million tokens (official Z.AI API price); Apertus 70B costs $0.82 input / $2.42 output per million tokens (median across 3 API providers). At a typical mix of three input tokens to one output token, that is $0.875 per million tokens for Qwen3-Next 80B-A3B Instruct versus $0.90 for GLM-4.5V (1× as much) and $1.22 for Apertus 70B (1.4× as much).
Which scores higher on benchmarks?
There is no independent benchmark that covers all three models yet. Apertus 70B has not been scored yet, GLM-4.5V has not been scored yet and Qwen3-Next 80B-A3B Instruct has not been scored yet.
Which is better for coding?
There are no published SWE-bench Verified results for Apertus 70B, GLM-4.5V and Qwen3-Next 80B-A3B Instruct yet, so there is no like-for-like coding score. Test both on a sample of your own tasks. All three support tool calling for agent workflows.
Which has the bigger context window?
Qwen3-Next 80B-A3B Instruct has the largest context window at 131,072 tokens, against 65,536 for Apertus 70B and 64,000 for GLM-4.5V. Maximum output per response: Apertus 70B up to 8,192, GLM-4.5V up to 16,384, Qwen3-Next 80B-A3B Instruct up to 32,768 tokens.
Which can read images, PDFs, audio or video?
Apertus 70B accepts text; GLM-4.5V accepts text, images and video; Qwen3-Next 80B-A3B Instruct accepts text. GLM-4.5V handles the widest range of inputs.
Are any of these open source?
Yes, all three publish their weights (Apache-2.0), so you can self-host them.
Which is newer?
Apertus 70B is the newest, released Sep 2, 2025. Qwen3-Next 80B-A3B Instruct came out Sep 2025; GLM-4.5V came out Aug 11, 2025. Knowledge cutoff: Apertus 70B Sep 2025, GLM-4.5V Apr 2025, Qwen3-Next 80B-A3B Instruct Apr 2025.
How do you decide the winner?
Each model gets a 0–100 score on capability (50%, independent benchmark results); price (25%, blended price per million tokens (3 input : 1 output), log scale); inputs & features (15%, image, PDF, audio and video input, tool calling, structured output and reasoning); context window (10%, maximum tokens per request, log scale). Dimensions missing for any model are dropped and the remaining weights rescaled, so every model is judged on the same evidence. Specs and prices come from public model listings and the labs’ own API pages; capability scores come from independent benchmark runs. Data updated Oct 4, 2026.
